/*******************************************************************************
Author: Jason Robey
Date: 9/4/2025
Purpose: This _README file describes the details of the replication package for 
	the following manuscript:
		Defining and Measuring Homicide Rates for Birth Cohorts:
		Methodological Challenges and Solutions 
		By Jason Robey and Matt Vogel
*******************************************************************************/

Stata .do files included:
0.0.ProgDef.do
3.1.1.Model_yearcat.do
3.1.2.Model_year.do
3.2.1.Simulation.do

Stata .dta files included:
2.1.1.merge_yearcat.dta
2.1.2.merge_year.dta 

The .do files are completed in numerical order from 0.ProgDef.do through 3.2.1.Simulation.do. 
Each .do file contains a block of code at the top that describes the purpose of the .do file. 
However, versions of the aggregated and merged .dta files are included in this replication package. 
The full analysis .do files (3.1.1.Model_yearcat.do & 3.1.2.Model_year.do) will run using the provided .dta files. 

General naming conventions:
Demographics (generally included either as a prefix or suffix to a variable name)
	a = all
	w = white
	b = black. 
	_cat = categorial age, year, or cohort variable

Cohort measurement strategies (defined in paper):
	pc = Pooled Cohort (PC)
	pc_lead = Pooled Cohort, Lead (PC-L)
	rc = Reconstructed Cohort (RC)
	rca1 = Reconstructed Cohort, No Overlap (RC-NO)
	rca2 = Reconstructed Cohort, No Overlap Alternate (RC-NOA)
	pc_both = Pooled Cohort, Both (PC-B)
	pc_my = Pooled Cohort, Mid Years (PC-MY)
	avg = 2-year rolling average for single-year cohorts (equivalent to RC) 

Criminal justice data: 
	har = homicide arrest rate
	hvr = homicide victimization rate 
	_log = log-transformed rate 

Population data: 
	pop = total population count
	pop_per_15_19 = relative cohort size (% of population aged 15-19)

Natality data: 
	nmar = non-marital birth rate 
  